Hi JZr,
I haven't gone through all the code changes in detail, but the overall idea
seems good - even if it probably will take some time to get used to seeing
errno_t everywhere (and it's a huge diff).
Nice idea with using the pointers to force a strict type check. It would be
nice if we could continuously type check errno_t. I don't think we should
include a dual definition of errno_t in the code, but I hope we could
eventually make the cstyle checker I am working on do this kind of
verification.
A few notes:
- kernel and boot should follow suit so that we keep a consistent style
everywhere
- will you be collapsing the changesets for integration? (probably good
idea)
Best regards,
Jiri
---------- Původní e-mail ----------
Od: Jiří Zárevúcky <[email protected]>
Komu: HelenOS development mailing list <[email protected]>
Datum: 11. 12. 2017 17:23:27
Předmět: [HelenOS-devel] errno_t
"Hello everyone,
I had promised to write an explanation for what I'm doing with error
codes, and then was too exhausted to follow through, but better late
than never.
Here's the explanation of the motivation and methodology:
https://le-jzr.github.io/helenos/c/errno/2017/12/11/abusing-type-checking-
for-fun-and-profit.html
If you don't have time to read it all and don't need a primer on the
status quo, you can skip to the "s/int/errno_t" heading.
The unpolished, unreviewed result of the work on uspace (beyond the
parts I already committed to mainline) is at
https://github.com/le-jzr/helenos/tree/errno
-- jzr
_______________________________________________
HelenOS-devel mailing list
[email protected]
http://lists.modry.cz/listinfo/helenos-devel
"
_______________________________________________
HelenOS-devel mailing list
[email protected]
http://lists.modry.cz/listinfo/helenos-devel
